Index
A
- accelerated computing developer program
- activation functions / Activation functions
- Adjusted Rand Index (ARI) / DNN performance analysis
- alternating least squares (ALS) / Model-based collaborative filtering
- Analog to Digital Converter (ADC) / Predictive models for clustering audio files
- Area Under Curve (AUC) / DNN performance analysis
- Artificial Neural Networks / Artificial Neural Networks
- attentional factorization machine (AFM) / Improved factorization machines for predictive analytics
- axon / Artificial Neural Networks
B
- Basic Linear Algebra Subroutines (BLAS) / Installing NVIDIA CUDA
- basic probability
- for predictive modeling / Basic probability for predictive modeling
- Bayes' rule / Bayes' rule
- Bazel
- URL, for installation / Installing TensorFlow from source
- bias neuron / Artificial Neural Networks
- biological neurons / Artificial Neural Networks
- BOW
- used, for predictive analytics (PA) / Using BOW for predictive analytics
- issues, defining / The problem definition
- dataset description / The dataset description and exploration
- dataset exploration / The dataset description and exploration
- spam prediction, LR used with TensorFlow / Spam prediction using LR and BOW with TensorFlow
- used, for spam prediction with TensorFlow / Spam prediction using LR and BOW with TensorFlow
- BRNN
- about / BRNNs
- used, for image classification / Using BRNN for image classification
C
- CBOW
- used, for word embedding / Using CBOW for word embedding and model building
- used, for model building / Using CBOW for word embedding and model building
- model building / CBOW model building
- reusing, for predicting sentiment / Reusing the CBOW for predicting sentiment
- central limit theorem (CLT)
- about / Central limit theorem
- skewness / Skewness and data distribution
- data distribution / Skewness and data distribution
- Chi-square independence test / Chi-square independence test
- Chi-square test / Chi-square tests
- clustering / Unsupervised learning and clustering
- clustering audio files
- predictive models / Predictive models for clustering audio files
- CNN-based predictive model
- for sentiment analysis / CNN-based predictive model for sentiment analysis
- movie and product review datasets, exploring / Exploring movie and product review datasets
- CNN, used for predictive analytics (PA) about reviews / Using CNN for predictive analytics about movie reviews
- CNN hyperparameters
- tuning / Tuning CNN hyperparameters
- CNN model
- for emotion recognition / CNN model for emotion recognition
- dataset description / Dataset description, Dataset description
- architecture design / CNN architecture design
- testing, on image / Testing the model on your own image
- used, for predictive analytics (PA) / Using complex CNN for predictive analytics
- CNN predictive model
- for image classification / CNN predictive model for image classification
- collaborative filtering (CF) / Factorization machines
- collaborative filtering approach
- for movie recommendations / Collaborative filtering approach for movie recommendations
- utility matrix / The utility matrix
- dataset description / Dataset description
- exploratory analysis, of dataset / Exploratory analysis of the dataset
- collaborative filtering approaches / Collaborative filtering approaches
- conditional entropy / Conditional entropy
- conditional probability
- about / Conditional probability
- chain rule / The chain rule of conditional probability
- content-based filtering approach / Content-based filtering approaches
- continuous bag-of-words / Continuous bag-of-words
- continuous skip-gram / Continuous skip-gram
- contrastive divergence / Restricted Boltzmann Machines
- conversion procedure
- reference link / Dataset description
- convolutional neural networks (CNN)
- about / CNNs and the drawbacks of regular DNNs
- architecture / CNN architecture
- convolutional operations
- about / Convolutional operations
- applying, in TensorFlow / Applying convolution operations in TensorFlow
- convolutional operations, parameters
- input / Applying convolution operations in TensorFlow
- filter / Applying convolution operations in TensorFlow
- strides / Applying convolution operations in TensorFlow
- padding / Applying convolution operations in TensorFlow
- use_cudnn_on_gpu / Applying convolution operations in TensorFlow
- data_format / Applying convolution operations in TensorFlow
- Cornell University (CU) / Exploring movie and product review datasets
- correlation / Covariance and correlation
- covariance / Covariance and correlation
- CUDA toolkit
- reference link / Installing NVIDIA CUDA
- cuDNN v5.1 library
- URL, for downloading / Installing NVIDIA cuDNN v5.1+
- curse of dimensionality / Working principles of a predictive model
D
- data, in TensorFlow
- reference link / Feeds and placeholders
- data distribution / Skewness and data distribution
- data model
- in TensorFlow / Data model in TensorFlow
- tensors / Tensors
- rank / Rank
- shape / Shape
- data type / Data type
- variables / Variables
- fetches / Fetches
- feeds / Feeds and placeholders
- placeholders / Feeds and placeholders
- dataset
- description / Description of the dataset
- URL, for downloading / Description of the dataset
- dataset description
- about / Dataset description, Dataset description
- bank-additional-full.csv / Dataset description
- bank-additional.csv / Dataset description
- bank-full.csv / Dataset description
- bank.csv / Dataset description
- ratings data / Ratings data
- movies data / Movies data
- user data / User data
- datasets, for DSCI 425
- reference link / Using K-means for predicting neighborhoods
- data type / Data type
- data value chain
- for creating decisions / Data value chain for making decisions
- DBN
- construction / Construction of a simple DBN
- unsupervised pretraining / Unsupervised Pretraining
- Decision trees (DTs) / Ensemble method for survival prediction: random forest
- deep belief networks
- about / Deep belief networks
- Restricted Boltzmann Machines / Restricted Boltzmann Machines
- DBN, construction / Construction of a simple DBN
- used, for predictive analytics (PA) / Using deep belief networks for predictive analytics
- reference link / Using deep belief networks for predictive analytics
- deep learning
- for predictive analytics (PA) / Deep learning for better predictive analytics
- Deep Neural Networks
- about / Deep Neural Networks
- architecture / DNN architectures
- performance analysis / DNN performance analysis
- dendrites / Artificial Neural Networks
- Distributed TensorFlow
- reference link / Tuning CNN hyperparameters
- DNN
- drawbacks / CNNs and the drawbacks of regular DNNs
E
- entropy
- about / Entropy
- Shannon entropy / Shannon entropy
- joint entropy / Joint entropy
- conditional entropy / Conditional entropy
- estimators
- about / Transformers and estimators
- estimator transformer / Estimator transformer
- expectation / Expectation
- exploratory analysis
- of dataset / Exploratory analysis of the dataset
F
- factorization machines
- for recommendation systems / Factorization machines for recommendation systems
- about / Factorization machines
- problem definition / Problem definition and formulation
- formulation / Problem definition and formulation
- dataset description / Dataset description
- preprocessing / Preprocessing
- FM model, implementing / Implementing an FM model
- improving, for predictive analytics (PA) / Improved factorization machines for predictive analytics
- neural factorization machines / Neural factorization machines
- Fandango
- reference link / Using linear regression for movie rating prediction
- Fast Fourier Transforms (FFT) / Installing NVIDIA CUDA
- feedforward neural network (FFNN) / DNN architectures
- fetches / Fetches
- fine-tuning DNN hyperparameters
- about / Fine-tuning DNN hyperparameters
- number of hidden layers / Number of hidden layers
- number of neurons per hidden layer / Number of neurons per hidden layer
- activation functions / Activation functions
- weight and biases initialization / Weight and biases initialization
- regularization / Regularization
- forwarding propagation step / Training an MLP
G
- Gated Recurrent Unit (GRU) / RNN architecture
- graphs / Vectors, matrices, and graphs
- GRU cell / GRU cell
H
- hidden layers / Multilayer perceptrons
- hybrid recommendation systems
- collaborative filtering / Hybrid recommendation systems
- content-based filtering / Hybrid recommendation systems
- hypothesis testing
- about / Hypothesis testing
- Chi-square test / Chi-square tests
- Chi-square independence test / Chi-square independence test
I
- information gain / Information gain
- information theory
- using, in predictive modeling / Using information theory in predictive modeling
- using / Using information theory
- using, in Python / Using information theory in Python
- input neurons / Artificial Neural Networks
- interquartile range (IQR) / Interquartile, range, and quartiles
J
- joint entropy / Joint entropy
- Jupyter Notebook
- reference link / Data type
K
- K-means
- used, for predictive analytics (PA) / Using K-means for predictive analytics
- working / How K-means works
- cluster assignment step / How K-means works
- update step / How K-means works
- used, for predicting neighborhoods / Using K-means for predicting neighborhoods
- betweenness / Using K-means for predicting neighborhoods
- withiness / Using K-means for predicting neighborhoods
- totwithiness / Using K-means for predicting neighborhoods
- Kaggle
- kNN
- used, for predictive analytics (PA) / Using kNN for predictive analytics
- working principles / Working principles of kNN
- -based predictive model, implementing / Implementing a kNN-based predictive model
L
- Latent Drichilet Allocation (LDA) / Using text analytics
- latent factors (LFs) / Model-based collaborative filtering
- least squares fitting (LSF) / A bit of linear algebra
- libcupti-dev library
- installing / Installing the libcupti-dev library
- linear algebra / Programming linear algebra
- about / A bit of linear algebra
- programming / Programming linear algebra
- SciPy / Programming linear algebra
- Linear algebra package (LAPACK) / Programming linear algebra
- NumPy / Programming linear algebra
- Pandas / Programming linear algebra
- linear independence / Span and linear independence
- linear regression
- about / Getting started with TensorFlow – linear regression and beyond, Linear regression - revisited
- source code / Source code for the linear regression
- problem statement / Problem statement
- used, for movie rating prediction / Using linear regression for movie rating prediction
- linear threshold unit (LTU) / Artificial Neural Networks
- log-likelihood function / Training an MLP
- Long Short-Term Memory (LSTM) / RNN architecture
- LSTM model evaluation / LSTM model evaluation
- LSTM model training / LSTM model training
- LSTM networks / LSTM networks
- LSTM predictive model / LSTM predictive model
- for sentiment analysis / An LSTM predictive model for sentiment analysis
- network design / Network design
- LSTM model training / LSTM model training
- TensorBoard, visualizing / Visualizing through TensorBoard
- LSTM model evaluation / LSTM model evaluation
M
- 1m MovieLens dataset
- reference link / Dataset description
- machine learning (ML) / Unsupervised learning and clustering
- marginal probability
- about / Marginal probability
- Markov Chain Monte Carlo (MCMC) / Restricted Boltzmann Machines
- Markov Random Fields (MRF) / Deep belief networks
- mathematical explanation
- reference link / Vectors
- matrices
- about / Vectors, matrices, and graphs, Matrices
- matrix addition / Matrix addition
- multiplying / Multiplying two matrices
- determinant of matrix, finding / Finding the determinant of a matrix
- transpose of a matrix, finding / Finding the transpose of a matrix
- simultaneous linear equations, solving / Solving simultaneous linear equations
- eigenvectors / Eigenvalues and eigenvectors
- eigenvalues / Eigenvalues and eigenvectors
- matrix factorization (FM) / Hybrid recommendation systems
- matrix subtraction / Matrix subtraction
- Mean Squared Error (MSE) / Implementing a kNN-based predictive model
- model-based collaborative filtering / Model-based collaborative filtering
- model evaluation / Model evaluation
- MovieLens data
- reference link / Training the model with available ratings
- movie recommendation engine
- implementing / Implementing a movie recommendation engine
- model, training with ratings / Training the model with available ratings
- saved model, inferencing / Inferencing the saved model
- user-item table, generating / Generating a user-item table
- movies, clustering / Clustering similar movies
- movie rating prediction, by users / Movie rating prediction by users
- top K movies, finding / Finding the top K movies
- top K movies, predicting / Predicting top K similar movies
- user-user similarity, computing / Computing the user-user similarity
- recommendation system, evaluating / Evaluating the recommendation system
- movie review data
- references / Exploring movie and product review datasets
- movie review dataset
- reference link / Using CBOW for word embedding and model building
- multiarmed bandit's predictive model
- developing / Developing a multiarmed bandit's predictive model
- multilayer perceptron
- about / Multilayer perceptrons
- MLP, training / Training an MLP
- MLP, used / Using MLPs
- preprocessing / Preprocessing
- TensorFlow implementation, of MLP / A TensorFlow implementation of MLP
- multilayer perceptrons
- used, for predictive analytics (PA) / Using multilayer perceptrons for predictive analytics
- dataset description / Dataset description
- mutual information / Mutual information
N
- N-grams / N-gram
- network design
- about / Network design
- embedding layer / Network design
- RNN layer / Network design
- softmax or sigmoid layer / Network design
- neural factorization machine (NFM) / Improved factorization machines for predictive analytics
- used, for movie recommendations / Using NFM for movie recommendations
- model training / Model training
- model evaluation / Model evaluation
- neural factorization machines
- about / Neural factorization machines
- dataset description / Dataset description
- Neural Networks (NNs) / Using kNN for predictive analytics
- NLP analytics pipelines
- about / NLP analytics pipelines
- text analytics, used / Using text analytics
- nonparametric model
- versus parametric model / Parametric versus nonparametric model
- nonparametric predictive models / Nonparametric predictive models
- notation
- number of hidden layers / Number of hidden layers
- number of neurons per hidden layer / Number of neurons per hidden layer
- numerical linear algebra (NLA) / Programming linear algebra
- NVIDIA CUDA
- installing / Installing NVIDIA CUDA
- NVIDIA cuDNN v5.1+
- installing / Installing NVIDIA cuDNN v5.1+
- NVIDIA Graph Analytics Library (nvGRAPH) / Installing NVIDIA CUDA
P
- padding operations / Pooling layer and padding operations
- subsampling operations, applying in TensorFlow / Applying subsampling operations in TensorFlow
- parametric model
- versus nonparametric model / Parametric versus nonparametric model
- parametric predictive models / Parametric predictive models
- perceptron / Artificial Neural Networks
- placeholder
- about / TensorFlow programming model
- data type / TensorFlow programming model
- shape / TensorFlow programming model
- name / TensorFlow programming model
- policy
- pooling layer / Pooling layer and padding operations
- population / Population and sample
- predictive analytics (PA)
- about / A basic introduction to predictive analytics, Why predictive analytics?
- predictive model, working principles / Working principles of a predictive model
- supervised learning / Supervised learning for predictive analytics
- K-means, used / Using K-means for predictive analytics
- kNN, used / Using kNN for predictive analytics
- BOW, used / Using BOW for predictive analytics
- deep learning / Deep learning for better predictive analytics
- deep belief networks, used / Using deep belief networks for predictive analytics
- Support Vector Machine (SVM) / CNN-based predictive model for sentiment analysis
- Clustering with mean-NN classification / CNN-based predictive model for sentiment analysis
- Naive Bayes / CNN-based predictive model for sentiment analysis
- One-hot (sequence) vectors / CNN-based predictive model for sentiment analysis
- N-grams / CNN-based predictive model for sentiment analysis
- predictive analytics tools
- in Python / Predictive analytics tools in Python
- predictive model
- developing, for time series data / Developing a predictive model for time series data
- dataset, description / Description of the dataset
- exploratory analysis / Preprocessing and exploratory analysis
- preprocessing / Preprocessing and exploratory analysis
- LSTM predictive model / LSTM predictive model
- model evaluation / Model evaluation
- predictive modeling
- information theory, using in / Using information theory in predictive modeling
- predictive models
- for clustering audio files / Predictive models for clustering audio files
- principal component analysis / Principal component analysis
- probability / Probability and the random variables
- probability density functions (PDF)
- about / Probability distributions
- probability distributions
- about / Probability distributions
- conditional independence / Independence and conditional independence
- independence / Independence and conditional independence
- probability mass functions (PMF)
- about / Probability distributions
- Python
- installing / Installing and getting started with Python
- obtaining / Installing and getting started with Python
- reference link / Installing and getting started with Python
- installing, on Windows / Installing on Windows
- installing, on Linux / Installing Python on Linux
- installing, on PIP / Installing and upgrading PIP (or PIP3)
- upgrading, on PIP / Installing and upgrading PIP (or PIP3)
- installing, on macOS / Installing Python on Mac OS
- packages, installing / Installing packages in Python
- about / Getting started with Python
- data types / Python data types
- string, used / Using strings in Python
- list, used / Using lists in Python
- tuple, used / Using tuples in Python
- dictionary, used / Using dictionary in Python
- set, used / Using sets in Python
- functions / Functions in Python
- classes / Classes in Python
- predictive analytics tools / Predictive analytics tools in Python
- information theory, using in / Using information theory in Python
- Python programming
- reference link / Classes in Python
R
- random numbers
- generating / Generating random numbers and setting the seed
- random sampling / Random sampling
- random variables / Probability and the random variables
- rank / Rank
- recommendation systems
- about / Recommendation systems
- collaborative filtering approaches / Collaborative filtering approaches
- cold start / Collaborative filtering approaches
- scalability / Collaborative filtering approaches
- sparsity / Collaborative filtering approaches
- content-based filtering approach / Content-based filtering approaches
- hybrid recommendation systems / Hybrid recommendation systems
- model-based collaborative filtering / Model-based collaborative filtering
- factorization machines / Factorization machines for recommendation systems
- in factorization machines / The cold start problem in recommendation systems
- rectified linear unit (ReLU) / DNN architectures
- regularization
- about / Regularization
- L2 regularization / Regularization
- max-norm constraints / Regularization
- dropout / Regularization
- reinforcement learning (RL)
- about / Reinforcement learning
- in predictive analytics / Reinforcement learning in predictive analytics
- requests package
- reference link / Implementing a kNN-based predictive model
- Restricted Boltzmann Machines / Restricted Boltzmann Machines
- Restricted Boltzmann Machines (RBM) / Deep belief networks
- RNN
- implementing, for spam prediction / Implementing an RNN for spam prediction
- RNN architecture
- about / RNN architecture, Contextual information and the architecture of RNNs
- contextual information / Contextual information and the architecture of RNNs
- BRNN / BRNNs
- LSTM networks / LSTM networks
- GRU cell / GRU cell
- Root Mean Squared Error (RMSE) / DNN performance analysis
- R script
- reference link / Using K-means for predicting neighborhoods
S
- sample
- about / Population and sample
- random sampling / Random sampling
- expectation / Expectation
- sandard deviation (SD) / Standard deviation and variance
- seed
- self-information
- about / Self-information
- mutual information / Mutual information
- sentiment analysis
- CNN-based predictive model / CNN-based predictive model for sentiment analysis
- LSTM predictive model / An LSTM predictive model for sentiment analysis
- Shannon entropy / Shannon entropy
- shape / Shape
- signals / Artificial Neural Networks
- singular value decomposition
- about / Singular value decomposition
- used, for data compression in predictive model / Data compression in a predictive model using SVD
- singular value decomposition (SVD) / Hybrid recommendation systems
- skewness / Skewness and data distribution
- Spam dataset
- reference link / The dataset description and exploration
- span / Span and linear independence
- squared error function / Training an MLP
- Staked Auto-Encoders (SAEs) / DNN architectures
- standard deviation (SD) / Standard deviation and variance
- standard transformer / Standard transformer
- statistical model
- about / Statistical models
- nonparametric model, versus parametric model / Parametric versus nonparametric model
- parametric predictive models / Parametric predictive models
- nonparametric predictive models / Nonparametric predictive models
- statistics
- used, in predictive modeling / Using statistics in predictive modeling
- population / Population and sample
- sample / Population and sample
- central limit theorem (CLT) / Central limit theorem
- sandard deviation (SD) / Standard deviation and variance
- variance / Standard deviation and variance
- interquartile range (IQR) / Interquartile, range, and quartiles
- quartiles / Interquartile, range, and quartiles
- hypothesis testing / Hypothesis testing
- stock price predictive model
- developing / Developing a stock price predictive model
- StopWordsRemover / StopWordsRemover
- supervised learning
- for predictive analytics (PA) / Supervised learning for predictive analytics
- Support Vector Machines (SVMs) / Using kNN for predictive analytics, Artificial Neural Networks
- synaptic terminals / Artificial Neural Networks
T
- telodendrian / Artificial Neural Networks
- TensorBoard
- about / TensorBoard
- working / How does TensorBoard work?
- visualizing / Visualizing through TensorBoard
- TensorFlow
- overview / General overview of TensorFlow
- faster computing / General overview of TensorFlow
- flexibility / General overview of TensorFlow
- portability / General overview of TensorFlow
- easy debugging / General overview of TensorFlow
- unified API / General overview of TensorFlow
- GPU computing, transparency / General overview of TensorFlow
- easy use / General overview of TensorFlow
- production scale / General overview of TensorFlow
- extensibility / General overview of TensorFlow
- supported / General overview of TensorFlow
- wide adoption / General overview of TensorFlow
- installing / Installing and configuring TensorFlow, Installing TensorFlow
- configuring / Installing and configuring TensorFlow
- installing, on Linux / Installing TensorFlow on Linux
- with CPU support / Installing TensorFlow on Linux
- with GPU support / Installing TensorFlow on Linux
- Python, installing / Installing Python and nVidia driver
- nVidia driver, installing / Installing Python and nVidia driver
- installing, with native pip / Installing TensorFlow with native pip
- installing, with virtualenv / Installing with virtualenv
- installing, from source / Installing TensorFlow from source
- installation, testing / Testing your TensorFlow installation
- computational graph / TensorFlow computational graph
- normal / TensorFlow computational graph
- special / TensorFlow computational graph
- tf.Operation objects / TensorFlow computational graph
- tf.Tensor object / TensorFlow computational graph
- variables / TensorFlow computational graph
- tensors / TensorFlow computational graph
- placeholders / TensorFlow computational graph
- session / TensorFlow computational graph
- programming model / TensorFlow programming model
- data model / Data model in TensorFlow
- Dataset API / Feeds and placeholders
- feeding / Feeds and placeholders
- reading, from files / Feeds and placeholders
- preloaded data / Feeds and placeholders
- convolutional operations, applying / Applying convolution operations in TensorFlow
- TensorFlow, on macOS
- URL, for installation / Installing and configuring TensorFlow
- TensorFlow, on Windows
- URL, for installation / Installing and configuring TensorFlow
- TensorFlow-based implementation of FM
- reference link / Problem definition and formulation
- TensorFlow contrib
- using / Using TensorFlow contrib
- reference link / Using TensorFlow contrib
- TensorFlow Implementation, of Neural Factorization Machine
- reference link / Using NFM for movie recommendations
- TensorFlow Python package
- reference link / Installing TensorFlow with native pip
- tensors / Vectors, matrices, and graphs
- about / Data model in TensorFlow, Tensors
- reference link / Tensors
- text analytics
- using / Using text analytics
- sentiment analysis / Using text analytics
- topic modelling / Using text analytics
- term frequency - inverse document frequency (TF-IDF) / Using text analytics
- named entity recognition / Using text analytics
- event extraction / Using text analytics
- TF-IDF model
- for predictive analytics (PA) / TF-IDF model for predictive analytics
- TF, computing / How to compute TF, IDF, and TFIDF?
- IDF, computing / How to compute TF, IDF, and TFIDF?
- TFIDF, computing / How to compute TF, IDF, and TFIDF?
- implementing, for spam prediction / Implementing a TF-IDF model for spam prediction
- tf.Tensor
- reference link / TensorFlow programming model
- time series data
- predictive model, developing / Developing a predictive model for time series data
- Titanic dataset
- about / Taking decisions based on data - Titanic example
- data value chain, for creating decisions / Data value chain for making decisions
- example / From disaster to decision – Titanic survival example
- titanic dataset
- about / From disaster to decision - Titanic example revisited
- exploratory analysis / An exploratory analysis of the Titanic dataset
- feature engineering / Feature engineering
- logistic regression, for survival prediction / Logistic regression for survival prediction
- TensorFlow contrib, using / Using TensorFlow contrib
- linear SVM, for survival prediction / Linear SVM for survival prediction
- ensemble method, for survival prediction / Ensemble method for survival prediction: random forest
- comparative analysis / A comparative analysis
- transformers
- about / Transformers and estimators
- standard transformer / Standard transformer
- estimator transformer / Estimator transformer
- StopWordsRemover / StopWordsRemover
- N-grams / N-gram
U
- UCI Machine Learning repository
- reference link / Dataset description
- unsupervised learning / Unsupervised learning and clustering
- unsupervised pretraining / Unsupervised Pretraining
- utility
- utility matrix / The utility matrix
V
- variables / Variables
- variance / Standard deviation and variance
- vectors / Vectors, matrices, and graphs, Vectors
W
- weight and biases initialization
- about / Weight and biases initialization
- zero initialization, avoiding / Weight and biases initialization
- small random numbers / Weight and biases initialization
- biases, initializing / Weight and biases initialization
- within-cluster sums of squares (WCSS) / DNN performance analysis
- Word2vec
- used, for sentiment analysis / Using Word2vec for sentiment analysis
- continuous bag-of-words / Continuous bag-of-words
- continuous skip-gram / Continuous skip-gram
- CBOW, used, for word embedding / Using CBOW for word embedding and model building
- CBOW, used, for model building / Using CBOW for word embedding and model building